perf: Cach hashing of categories

Author: fjetter

doc/source/whatsnew/v1.3.0.rst

@@ -388,6 +388,7 @@ Performance improvements
 - Performance improvement in :class:`Styler` where render times are more than 50% reduced (:issue:`39972` :issue:`39952`)
 - Performance improvement in :meth:`core.window.ewm.ExponentialMovingWindow.mean` with ``times`` (:issue:`39784`)
 - Performance improvement in :meth:`.GroupBy.apply` when requiring the python fallback implementation (:issue:`40176`)
+- Performance improvement for concatenation of data with type :class:`CategoricalDtype` (:issue:`40193`)
 
 .. ---------------------------------------------------------------------------
 

pandas/core/dtypes/dtypes.py

@@ -343,6 +343,7 @@ def _finalize(self, categories, ordered: Ordered, fastpath: bool = False) -> Non
 
         self._categories = categories
         self._ordered = ordered
+        self._hash_cache = -10  # Sentinel. -1 and -2 are used below with some semantics
 
     def __setstate__(self, state: MutableMapping[str_type, Any]) -> None:
         # for pickle compat. __get_state__ is defined in the
@@ -352,6 +353,8 @@ def __setstate__(self, state: MutableMapping[str_type, Any]) -> None:
         self._ordered = state.pop("ordered", False)
 
     def __hash__(self) -> int:
+        if self._hash_cache >= 0:
+            return self._hash_cache
         # _hash_categories returns a uint64, so use the negative
         # space for when we have unknown categories to avoid a conflict
         if self.categories is None:
@@ -360,7 +363,8 @@ def __hash__(self) -> int:
             else:
                 return -2
         # We *do* want to include the real self.ordered here
-        return int(self._hash_categories(self.categories, self.ordered))
+        self._hash_cache = int(self._hash_categories(self.categories, self.ordered))
+        return self._hash_cache
 
     def __eq__(self, other: Any) -> bool:
         """